Reconstruction of a surface electrocardiogram based upon an endocardial electrogram

Abstract

The reconstruction of a surface electrocardiogram based upon an endocardial electrogram. This method includes: (a) acquisition ( 10 ) of a plurality of endocardial electrogram signals (EGM) through a plurality of endocardial leads defined based upon endocardial electrodes; (b) calculation ( 12 ), by combining the endocardial electrogram (EGM) signals acquired at step (a), of the corresponding endocardial vectogram (VGM); (c) angular rescaling ( 14 ) of the orthonormalized mark of the endocardial vectogram (VGM) with that of the surface vectocardiogram (VCG); (d) estimation ( 16 ), based upon the endocardial vectogram (VGM) calculated at step (b), of a reconstructed surface vectocardiogram (VCGreconstructed), and (e) calculation ( 18 ) of the surface electrocardiogram (ECG) corresponding to said reconstructed surface vectocardiogram (VCGreconstructed).

1. A process for processing signals representative of cardiac myocardium depolarization potentials, said signals being collected by a plurality of endocardial electrodes of an active implantable medical device such as an implantable pacemaker, device for resynchronization, cardioversion and/or defibrillation, the process being performed by one of: the active implantable medical device, an external programmer, data server, the process comprising the steps of:
 a) acquiring a plurality of endocardial electrogram signals (EGM) through a plurality of leads, each lead of the plurality of leads being collected between at least two pairs of endocardial electrodes selected from the plurality of endocardial electrodes; 
 b) calculating, through a combination of the plurality of endocardial electrogram signals (FGM) acquired at step a), a corresponding endocardial vectogram (VGM); 
 d) estimating, based upon the endocardial vectogram calculated at step b), a reconstructed surface vectocardiogram (VCGreconstructed); and 
 e) reconstructing surface electrocardiogram signals (ECG) corresponding to said reconstructed surface vectocardiogram (VCGreconstructed). 
 
     
     
       2. The process of  claim 1 , wherein step b) of calculating the endocardial vectogram (VGM) further comprises performing an orthogonalization process. 
     
     
       3. The process of  claim 2 , wherein performing said orthogonalization process further comprises performing a Karhuen-Loeve transform of said combination of EGM signals. 
     
     
       4. The process of  claim 1 , further comprising, between steps b) and d):
 c) angular rescaling the orthonormalized mark of the endocardial vectogram (VGM) upon that of the surface vectocardiogram (VCG). 
 
     
     
       5. The process of  claim 4 , further comprising a preliminary step of calibration comprising determining parameters of said angular rescaling. 
     
     
       6. The process of  claim 5 , wherein determining angular rescaling parameters further comprising the steps of:
 i) obtaining a set of reference data through simultaneous acquisition of endocardial electrogram signals (EGM) and surface electrocardiogram signals (ECG); 
 ii) calculating, through a combination of the surface electrocardiogram signals (ECG) acquired at step i), a corresponding surface vectocardiogram (VCGreal); 
 iii) calculating, through a combination of endocardial electrogram signals (EGM) acquired at step i), a corresponding endocardial vectogram (VGM); 
 iv) angular rescaling the orthonormalized mark of the endocardial vectogram (VGM) upon that of the surface vectocardiogram (VCGreal); 
 v) estimating, based upon the endocardial vectogram (VGM) calculated at step iii), of a reconstructed surface vectocardiogram (VCGreconstructed); and 
 vi) adjusting the angular rescaling parameters of step iv) so as to minimize a deviation between the surface vectocardiogram (VCGreal) calculated at step ii) and the reconstructed surface vectocardiogram (VCGreconstructed) calculated at step v). 
 
     
     
       7. The process of  claim 6 , wherein said deviation to be minimized is the root-mean-square deviation. 
     
     
       8. The process of  claim 6 , further comprising providing an adaptive neural network for a determination of said angular rescaling parameters. 
     
     
       9. The process of  claim 1 , wherein, the step d) of estimating the reconstructed surface vectocardiogram (VCGreconstructed) further comprises applying a non-linear filtering to the endocardial vectogram calculated at step b). 
     
     
       10. The process of  claim 9 , further comprising providing an adaptive neural network to implement said non-linear filtering, 
     
     
       11. The process of  claim 9 , further comprising a preliminary calibration step comprising determining parameters of said non-linear filtering. 
     
     
       12. The process of  claim 11 , wherein determining non linear filtering parameters further comprising the steps of:
 i) obtaining a set of reference data through simultaneous acquisition of endocardial electrogram signals (EGM) and surface electrocardiogram signals (ECG); 
 ii) calculating, through a combination of the surface electrocardiogram signals (ECG) acquired at step i), a corresponding surface vectocardiogram (VCGreal); 
 iii) calculating, through combination of endocardial electrogram signals (EGM) acquired at step i), a corresponding endocardial vectogram (VGM); 
 iv) conducting a possible angular rescaling of an orthonormalized mark of the endocardial vectogram (VGM) upon that of the surface vectocardiogram (VCGreal); 
 v) estimating, through applying non-linear filtering to the endocardial vectogram calculated at step iii), a reconstructed surface vectocardiogram (VCGreconstructed); and 
 vi) adjusting the non-linear filtering parameters of step v) so as to minimize a deviation between the surface vectocardiogram (VCGreal) calculated at step ii) and the reconstructed surface vectocardiogram (VCGreconstructed) calculated at step v). 
 
     
     
       13. The process of  claim 9 , further comprising providing to said non-linear filtering as an input, at least one parameter selected from among: at respiratory signal, information on the position of the endocardial sensing electrodes, the P, QRS or T phase of the cardiac cycle; and a signal representative of intracardiac impedance. 
     
     
       14. The process of  claim 1 , wherein said intracardiac signals are collected using at least one of: right ventricular distal and/or proximal electrode, right atrial distal and/or proximal electrode, left ventricular distal and/or proximal electrode, ventricular or atrial defibrillation coil, and supra-ventricular defibrillation coil. 
     
     
       15. The process of  claim 1 , wherein the plurality of endocardial electrodes of the active implantable medical device comprises two to four endocardial electrodes implanted to a patient. 
     
     
       16. Apparatus for processing signals representative of cardiac myocardium depolarization potentials, comprising:
 a) means for acquiring a plurality of endocardial electrogram signals (EGM) representative of a plurality of leads; 
 b) means for calculating, based on a combination of said acquired plurality of endocardial electrogram signals (EGM), a corresponding endocardial vectogram (VGM); 
 c) means for estimating, based upon the calculated endocardial vectogram, a reconstructed surface vectocardiogram (VCGreconstructed); and 
 d) means for calculating surface electrocardiogram signals (ECG) corresponding to said reconstructed surface vectocardiogram (VCGreconstructed). 
 
     
     
       17. The apparatus of  claim 16 , wherein said means for calcu 1 ating the endocardial vectogram (VGM) further comprises means for performing an orthogonalization process. 
     
     
       18. The apparatus of  claim 17 , wherein said means for performing said orthogonalization process further comprises means for performing a Karhuen-Loeve transform of said combination of EGM signals. 
     
     
       19. The apparatus of  claim 16 , further comprising:
 means for angular rescaling an orthonormalized mark of the endocardial vectogram (VGM) upon that of the surface vectocardiogram (VCG) prior to estimating said reconstructed surface vectocardiogram. 
 
     
     
       20. The apparatus of  claim 19 , further comprising means for determining parameters of said angular rescaling during a preliminary calibration. 
     
     
       21. The apparatus of  claim 16 , wherein, the means for, estimating the reconstructed surface vectocardiogram (VCGreconstructed) further comprises means for applying a non-linear filtering to the calculated endocardial vectogram, said means for nonlinear filtering further comprising a neural network. 
     
     
       22. The apparatus of  claim 16 , wherein the plurality of endocardial electrodes of the active implantable medical device comprises two to four endocardial electrodes implanted to a patient. 
     
     
       23. A non-transitory software control module for processing signals representative of cardiac myocardium depolarization potentials, comprising:
 a) a first instruction set for acquiring a plurality of endocardial electrogram signals (EGM) representative of a plurality of leads; 
 b) a second instruction set for calculating, based on a combination of said acquired endocardial electrogram signals (EGM), a corresponding endocardial vectogram (VGM); 
 c) a third instruction set for estimating, based upon the calculated endocardial vectogram, a reconstructed surface vectocardiogram (VCGreconstructed); and 
 d) a fourth instruction set for calculating surface electrocardiogram signals (ECG) corresponding to said reconstructed surface vectocardiogram (VCGreconstructed). 
 
     
     
       24. The non-transitory software control module of  claim 23 , wherein said second instruction set further comprises instructions performing an orthogonalization process by performing a Karhuen-Loeve transform of said combination of EGM signals. 
     
     
       25. The non-transitory software control module of  claim 23 , further comprising:
 an instruction set for angular rescaling an orthonormalized mark of the endocardial vectogram (VGM) upon that of the surface vectocardiogram (VCG) prior to estimating said reconstructed surface vectocardiogram. 
 
     
     
       26. The non-transitory software control module of  claim 23 , further comprising an instruction set for determining parameters of said angular rescaling during a preliminary calibration. 
     
     
       27. The non-transitory software control module of  claim 26 , wherein the instruction set for determining angular rescaling parameters further comprise instructions for:
 i) obtaining a set of reference data through simultaneous acquisition of endocardial electrogram signals (EGM) and surface electrocardiogram signals (ECG); 
 ii) calculating, through a combination of the surface electrocardiogram signals (ECG) acquired at step i), a corresponding surface vectocardiogram (VCGreal); 
 iii) calculating, through a combination of endocardial electrogram signals (EGM) acquired at step i), a corresponding endocardial vectogram (VGM); 
 iv) angular rescaling the orthonormalized mark of the endocardial vectogram (VGM) upon that of the surface vectocardiogram (VCGreal); 
 v) estimating, based upon the endocardial vectogram (VGM) calculated at step iii), of a reconstructed surface vectocardiogram (VCGreconstructed); and 
 vi) adjusting the angular rescaling parameters of step iv) so as to minimize the deviation between the surface vectocardiogram (VCGreal) calculated at step ii) and the reconstructed surface vectocardiogram (VCGreconstructed) calculated at step v). 
 
     
     
       28. The non-transitory software module of  claim 23 , wherein the plurality of endocardial electrodes of the active implantable medical device comprises two to four endocardial electrodes implanted to a patient.